Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

RE: spectral response acceleration definition

[Subject Prev][Subject Next][Thread Prev][Thread Next]
I agree with A. Daneshvar's response and will add some additional explanation. 
I would say that "spectral" has more to do with variation in response based on the natural period of the structure.  The response spectrum curve represents how a single degree of freedom system with any given natural period would respond to a given ground motion.  The short period spectral response represents the peak response or maximum response (i.e., due to resonance of the structure natural period and the applied earthquake). The one second spectral response simply represents a point on the curve in the long period range of response.  With these two points defined, a full response spectrum curve can be constructed to cover all periods, using the code defined equations. 

William C. Sherman, PE
(Bill Sherman)
CDM, Denver, CO
Phone: 303-298-1311
Fax: 303-293-8236
email: shermanwc(--nospam--at)


From: daneshvar.ahmad [mailto:daneshvar(--nospam--at)]
Sent: Tuesday, January 03, 2006 11:02 PM
To: seaint(--nospam--at)
Subject: RE: spectral response acceleration definition

Dear Bruce,


Hi, I believe the first and third paragraph is completely right, But there is a misunderstanding in the second one. My comment is as follow:


We need short (0.2 sec) and long (1.0 sec) period “spectral response acceleration” to construct “response spectrum curve” accordingly. Short period spectral response acceleration Ss is similar to Ca in 97 UBC.( as a measure of constant acceleration part of response spectrum curve for relatively rigid structures with short period), and long period spectral acceleration S1 is similar to Cv in 97 UBC (as a measure of constant velocity part of response spectrum curve for relatively flexible structures with long period). Therefore, after constructing of response spectrum curve the spectral response acceleration for other periods to be extracted from this curve. Code permits to evaluate the Ss and S1 for location between the contours of the maximum considered earthquake ground motion’s maps by interpolation.


Have a good presentation!

A. Daneshvar
Civil Dept., PIDEC

From: Bruce Holcomb [mailto:bholcomb(--nospam--at)]
Sent: Tuesday, January 03, 2006 7:31 PM
To: seaint(--nospam--at)
Subject: spectral response acceleration definition


I have been having trouble with the term “spectral” in “spectral response acceleration”, so I’ve been doing some studying.  Can someone verify my findings below?  Or set me straight?  I’m supposed to give a presentation at my office and I want to make sure I have it straight before I try to explain to others. 


The “spectral response acceleration” plotted in the “Maximum Considered Earthquake Ground Motion” maps in the IBC is an acceleration with an annual probability of exceedance of 2% in 50 years (return period of 2500 years) and is for a site class “B” soil.  The spectral response acceleration is derived from many possible earthquakes of various magnitudes at various distances from the site.


The spectral response acceleration is plotted for period of 0.2 seconds and 1.0 seconds because these two periods encompass most building periods.  Periods between these values (and to a certain extent beyond these values) can be interpolated (extrapolated).


The reason the “design spectral response acceleration” is 2/3 of the “site adjusted spectral response acceleration” is that the lower bound estimate of the margin against collapse is a factor of 1.5 (approximate… judged by experience).  Therefore, the “design” earthquake ground motion is 1/1.5 (2/3) of the maximum considered earthquake ground motion because we want to prevent collapse, but the construction cost would be too great, if we designed to prevent all damage to a structure.